Presenter: Lidong Song, Ph.D. Student

VIEW THE RECORDING HERE.

It is a common practice for utilities to down-sample smart meter measurements from high resolution (e.g. 1-min or 1-sec) to low resolution (e.g. 15-, 30- or 60-min) to lower the data transmission and storage cost. However, down-sampling can remove high-frequency components from time-series load profiles, making them unsuitable for in-depth studies such as quasi-static power flow analysis or non-intrusive load monitoring (NILM). This presentation will review a Generative Adversarial Network (GAN) based load profile super-resolution (LPSR) framework for restoring high-frequency components lost through the smoothing effect of the down-sampling process. This method outperforms the traditional Mean Square Error (MSE) based super-resolution methods. Improving LPSR in this way can reduce utility costs while still leveraging the benefits of smart meter measurements.